Transferred expectations of human presence: Folk theories among older adults who are inexperienced users of online services

Abstract

This study analyses the expectations of older adults who are inexperienced users of online media and services, examining their sense-making processes when using the internet for informational and practical purposes. Research on older users often focuses on access and abilities, but this study instead explores older adults’ expectations of what it means to interact online. We apply a ‘folk theory’ framework to illuminate underlying perceptions that guide behaviours, by asking which folk theories older adults draw on to make sense of their experiences with the internet. The empirical data originates from qualitative in-depth interviews and participant observation sessions with 25 people aged 65–98 years in Norway. We identify four interconnected folk theories under the shared theme of transferring expectations from the offline world: expecting human involvement, expecting visibility, lack of a human safety net, and human limitations and social conventions. Our analysis shows how such folk theories inform user decisions, including hindrances and problem-solving, as older adults adapt to digital services in everyday life.

Introduction

As societies become more digital, older adults must increase their use of online services relevant to their economic, political, cultural, and private lives, which may lead to deeper digital divides (Friemel, 2016). An increasing number of services required in everyday life are primarily found online, including banking, ordering home deliveries, booking appointments, buying tickets, contacting healthcare providers, communicating with public offices, or accessing user support through chatbots or digital contact forms. However, societies are ‘ageing and mediatising at the same time’ (Givskov and Deuze, 2018: 399). While the idea of digitalisation is often associated with benefits such as prolonged independence, reduced social isolation and improved cognitive abilities (Niehaves and Plattfaut, 2014), older adults may not actually feel aided by digital services (Hänninen et al., 2023; Lehtinen, 2023). The rhetoric surrounding older adults and technology could influence how they perceive their digital literacies, creating an atmosphere of self-doubt or reducing willingness to develop digital skills (Mitzner et al., 2016; Schreurs et al., 2017). To examine barriers to digital inclusion, the everyday internet experiences of older users are an important area of analysis.

A fundamental question is why older and younger internet users might perceive the digital world very differently. As underlined by a generational approach to media use, older adults are likely to have established non-digital media habits (Bolin, 2017; Ivan et al., 2020), as part of lived experiences from a time when society was less digital. Their understandings of their lifeworlds, and of how to build a stock of knowledge for navigating everyday life (Schutz and Luckmann, 1973), could involve distinct practices and beliefs that play into how they perceive the internet. We suggest that this can be analysed through a ‘folk theory’ framework, starting from the premise that online environments are mostly navigated using experiences and predictions rather than in-depth technical knowledge. French and Hancock (2017) describe how ‘folk theories’ guide navigation of complex systems, helping people reason and predict how systems work. The folk theory approach has been applied in studies of digital media (Siles et al., 2020; Toff and Nielsen, 2018; Ytre-Arne and Moe, 2021) to shed light on users’ conceptions of how online environments function. While folk theories are not a particularity of older adults, their theories are likely to be different to those of younger users, related to differences in lived experiences. We therefore draw on a folk theory framework to investigate how older adults theorise their everyday encounters with online services.

We pose the following research question: Which folk theories do older adults draw on when making sense of their experiences with the internet, particularly online services? Through analysis of qualitative in-depth interviews and observation sessions with older adults in Norway, we argue that transferred expectations of human presence, originating from the offline world play a key role in guiding behaviours online. Our study contributes to the research literature on everyday experiences of older adults in the new media environment, by enhancing our understanding of the interpretations older adults assign to their interactions with the digital world. We conclude by discussing how the folk theories of older adults might have implications for the further development of digital literacies among this group, pointing to hindrances in building critical awareness and learning from experience.

Literature review: older adults online

Although less studied than adolescents, older adults represent a demographic group that has gained increasing attention in digital media research. Many studies focus on how internet use disparity leads to inequality, often corresponding to sociodemographic dimensions (Ivan et al., 2020; Korupp and Szydlik, 2005), and the concepts digital divides and digital exclusion/inclusion are linked to uneven distribution of resources and life chances (Van Dijk and Hacker, 2003). Regarding older adults, the term grey divide describes how internet use is skewed towards potential exclusion of this group (Friemel, 2016; Morris and Brading, 2007). As indicated by a focus on digital inclusion, the internet is often approached as a solution for older adults (Kebede et al., 2022), as demonstrated in studies on e-health (Welch et al., 2016), digital well-being (Sen et al., 2022), or social media use (Khosravi et al., 2016). However, some scholars have observed a paradox in older adults’ perceptions of online media and services, indicating limited engagement and interest (Sourbati, 2009; Van Deursen and Helsper, 2015). Despite some characterisations of older adults as late adopters of digital technology (Quan-Haase et al., 2018), research shows they are a heterogenous group of Internet users (Hargittai et al., 2019), more varied in adoption, usage, and skills than younger cohorts (Hänninen et al., 2021).

Two different reviews both conclude with the need to better understand the internet in the lives of older adults. Hunsaker and Hargittai (2018) reviewed quantitative studies of internet use among older adults and found emphasis on access and skills, concluding that much work remains to understand how the Internet positively or negatively influences older adults’ lives. Givskov and Deuze (2018) noted blind spots in current research on older people in the new media environment and argued for the need to explore daily media use and social categories in more depth, and move beyond questions of mere access.

Some qualitative contributions have explored such directions, indicating that older users often compare online communication to offline ideals. One study found Janus-faced conceptions of technology, with the disappearance of a human touch as a drawback (Pirhonen et al., 2020). Another found refusal to use digital devices based on social representations of such technology as ‘useless and risky’, potentially threatening appreciated values and lifestyles (Hakkarainen, 2012). Studies of older adults’ social media use underline questions of media richness and sociability (Jung et al., 2017; Lüders and Brandtzæg, 2017). Another recent strand in qualitative studies focuses on assistance from so-called ‘warm experts’, such as friends and family (Martínez and Olsson, 2022), while studies of appropriation processes underline the relevance of relations and practices in everyday life (Martínez, 2022). These strands of research all indicate that experiences from offline situations play into older adults’ sense-making when using digital media and services. We propose to investigate this further through the notion of folk theories, which also connects to debates on different dimensions of digital literacies.

Digital literacies

Digital literacy is a frequently discussed concept which is difficult to define, as it encompasses specific technical skills but also broader social practices, knowledge, and values (Iordache and Mariën, 2017; Pangrazio et al., 2020). Digital literacy can be defined as a social practice that involves meaning-making using a range of digital technologies (Sefton-Green et al., 2017). This definition involves cultural competence, which is particularly relevant for this study because it describes the ability to understand cultural signs embodied in meaning-making (Green, 1988). Moreover, digital literacy involves digital technologies and non-digital practices, crossing online-offline and material-immaterial boundaries (Sefton-Green et al., 2017). Livingstone (2011) characterised internet literacy as ‘culturally regulated competencies encompassing that which is normatively valued and that which is disapproved or transgressive’ (p. 106). While improving literacies is an objective of education and policy, literacy scholars are also interested in how digital media are used by individuals in everyday life, and how digital literacy relates to social, technological, and economic changes (Sefton-Green et al., 2017). The notion of ‘critical digital literacy’ further emphasises abilities for critical reflections on societal power structures pertaining to digital media, although scholars underline this should be balanced with attention to how individuals perceive and use digital technologies, instead of top-down prescriptions (Pangrazio, 2016; Pötzsch, 2019). Conceptualizations of digital literacies among social groups, such as a recent study of parents’ algorithmic literacies (Das, 2023), include the following dimensions: awareness, technical competencies, critical capacities, and championing of users’ best interests. The last dimension particularly speaks to how normative aspects of literacies should be connected to what is valuable to the users themselves.

As noted above, research on older adults as internet users has moved from a focus on access to increasingly emphasise skills (Hargittai et al., 2019; Van Deursen and Helsper, 2015). Lack of skills is a barrier to using digital technology and online services, and terms such as ‘digital skill divide’ or ‘second digital divide’ describe the challenges faced by people with limited knowledge of the internet and its applications and platforms (Hargittai, 2001). Research has also expanded the scope from skills to social practices and values: Older adults often choose not to use digital services owing to concerns about decreased human interaction and privacy (Lehtinen, 2023). In addition, even older adults who find digital services valuable and use them regularly might not necessarily like or prefer them (Hänninen et al., 2023). Attitudes towards the internet, as well as social and psychological barriers, are found to be important hindrances towards the potential benefits that older adults might gain from internet use (Van Deursen and Helsper, 2015). Schreurs et al. (2017) describe a ‘digital literacy paradox’ in which lack of digital skills, combined with negative media discourse and lacking social support, make older adults apprehensive about using digital media and thereby prevents them from gaining new experiences that are central to advancing literacies. They found that older adults tended to develop skills for specific activities, which limits their exposure to various digital services that could benefit their lives, referred to as the ‘digital skill bubble’. This is in line with studies finding selective online repertoires among older adults, using the Internet for specific functions such as news and online banking (Olsson et al., 2019). In addition, it can be challenging for older adults to gain the skills needed and improve literacies, as they do not always have the necessary support (Geerts et al., 2023).

Our study explores personal digital narratives and everyday experiences, which Pangrazio (2016) claims is needed to advance debates on critical literacies. We apply a folk theory framework to older adults’ experiences online, shedding light on their underlying perceptions and values. The conceptions brought to light through this approach could untangle paradoxes regarding digital literacies among older adults, explaining hindrances towards developing skills and critical awareness further—and to using digital technologies for the users’ own interests. These are point we will return to in our concluding discussion.

Methods

Norway is a digitally advanced society with an ageing population and a political push towards digital inclusion, but studies on older users’ internet experiences are scarce (Lüders and Brandtzæg, 2017; Syvertsen, 2011). ‘Everybody online’ has been an ambition for the development of public services in Norway, as in other Organisation for Economic Co-operation and Development (OECD) countries, mixed with a strong digital infrastructure (Syvertsen, 2020). Norwegian public services continue to accommodate non-digital solutions while increasingly implementing ‘digital first choice’ policies: For instance, the tax office requires those who want a non-digital option to request it online. However, a report from a government agency published in 2021 found that 600,000 Norwegians (11% of the population) were considered weak or non-digital users (Bjønness et al., 2021). Almost 50% of people over 60 years of age belong to these categories, and the percentage increases in the higher age brackets (Bjønness et al., 2021).

Our study, conducted in 2022, focuses on older adults who use the internet but are relatively inexperienced online. We operationalise relative inexperience through self-reported conceptions of facing challenges or having limited competencies while using self-selected online services in everyday life. The participants were recruited through public libraries and meeting places for seniors, where the project was presented. Those who recognised themselves as facing challenges with digital services in everyday settings were encouraged to volunteer for the study, while older adults who had zero or very extensive digital experience were not invited. Twenty-five retired adults aged 65–98 years participated in the study. In Norway, the general retirement age is 67 years, with early retirement available from 62 years. The participant group was balanced in terms of gender and included people with different educational levels (see Supplemental Material Appendix I). Differences in age and educational background implied that some had used computers for specific tasks when working, while others had recently acquired their first digital device.

Responding to calls for further research on the everyday experiences of older adults in the new media environment (Givskov and Deuze, 2018), our study employed qualitative ethnographic methods with in-depth interviews and participatory observation sessions in domestic settings.

The researcher visited the participants twice in their homes, which facilitated rich descriptions of digital technology in everyday life: Some had a laptop collecting dust on a bookshelf, others a desktop computer that occupied most of the dining table, or a smartphone or tablet or mixed setup of devices. During the second meeting, the participants were more tuned into what the researcher wanted to discuss, and many had ‘saved’ tasks they wanted to do online, allowing the researcher to reflect on insights from the first meeting and adjust for misunderstandings. This was particularly important, as participants sometimes lacked a ‘digital vocabulary’ to explain their online activity. Developing such a shared vocabulary was part of the process, after earlier pilot interviews had identified this key challenge. For instance, the term ‘internet’ was confusing and misdirected the conversation, while participants found it easier to talk about specific devices. Moreover, the participants were interested in practical and informational purposes, such as buying tickets or checking weather forecasts, and less interested in social media. Discussing and demonstrating online services chosen by the participants was very fruitful. Therefore, this study operates with flexible and user-dependent conceptions of ‘digital services’, ‘digital media’, ‘being online’, and ‘using digital technology’.

The interviews began with a ‘day in the life’ segment (Del Rio Carral, 2014) to obtain a sense of the participants’ daily lives and the devices they used. Subsequently, participants were asked about experiences with online banking or email, if these had not already been mentioned. Towards the end of the interview, questions about the participants’ thoughts on digital technology in society were included (see Supplemental Material Appendix II). Initial interviews were followed by participatory observations after 1 week, using the think-aloud method (Wirth et al., 2004). The researcher asked the participants to talk through something they usually did online, and then to visit a webpage new to them, such as Google or nrk.no (Norwegian public broadcaster), to examine how unfamiliar tasks differed from established habits. The process ensured flexible in-depth engagement, cleared up misunderstandings, and provided reflections on challenges and motivations. All interviews and participatory observations were recorded and transcribed.

The study followed the research ethics guidelines of the National Committee for Social Sciences and Humanities (NESH) in Norway and institutional systems for data protection in research projects. The study was assessed by the Norwegian Centre for Research Data (notification number 980638). All participants provided written, informed consent to participate. Participant information was kept anonymous using pseudonyms. Observation sessions were conducted without video or screen capture, and in a manner that ensured that participants did not need to share passwords or browsing histories. If the participants shared personal information during interviews, this was retracted from transcripts.

Folk theories as a framework

Our analysis draws on a folk theory framework. The concept of folk theories (also known as intuitive or common-sense theories) highlights people’s underlying conceptions that guide daily sense-making processes. French and Hancock (2017) described these understandings of the social world as implicit collections of beliefs, often formed through recurring everyday experiences or observations. Folk theories may be fleeting, inconsistent, and based on limited or misleading knowledge. Nonetheless, these theories guide individuals’ behaviours and expectations (Gelman and Legare, 2011). DeVito et al. (2017) distinguished between abstract folk theories, often guided by feelings towards phenomena, and operational folk theories, which are specific interpretations of functionality.

In research on digital media use, folk theory approaches are valuable to explore not only what people do or think they do ‘but what it means to them’ (Toff and Nielsen, 2018: 639). Folk theories have been defined as ‘the culturally available symbolic resources that people use to make sense of their own media and information practices’ (Toff and Nielsen, 2018: 637). Studies have analysed folk theories of algorithms (Siles et al., 2020; Ytre-Arne and Moe, 2021), information discovery (Toff and Nielsen, 2018), and newsfeed curation (Eslami et al., 2016). Similar approaches analyse algorithmic imaginaries and decodings (Bucher, 2017; Lomborg and Kapsch, 2020).

The folk theory approach acknowledges the value of perceptions that help individuals navigate the world (French and Hancock, 2017) without requiring profound or scientific knowledge of the phenomenon in question. In our study, folk theories concern how online environments function, and are rooted in everyday experiences, as described by Toff and Nielsen (2018). We use a folk theory framework to elucidate how older adults—with common-sense knowledge, non-digital media habits, and experiences from a less online world—perceive digital environments when using online services in everyday life.

The analysis involved several steps. Already at the data collection stage, we found that participants frequently talked about ‘them’, ‘they’, and ‘people’ as part of highly automated online services. Therefore, we identified all passages in the transcribed interviews which referred to another person. We tried to distinguish how different expressions of human presence would have appeared to our participants, connected to their accounts of experiences, looking for underlying understandings. Following other studies (Toff and Nielsen, 2018; Ytre-Arne and Moe, 2021), we formulated and reformulated potential theories, which we discussed based on the materials and previous research on digital literacies among older adults. Moreover, we drew on scholarship on human interpretations of digital communication, including anthropomorphism (Epley et al., 2007), metaphors as a guide for abstract concepts (Lakoff and Johnson, 1980), and the conceptualisation of the internet as a place (Markham, 2003), to signal how the participants’ folk theories correspond to or counteract other understandings of digital environments.

Analysis: four folk theories

Our analysis suggests four folk theories concerning different aspects of expecting human presence when using online services. All theories share the same overarching narrative, with connections to offline human communication and how the physical world functions, but otherwise emphasise different aspects of what it means to interact online. These theories are expectations of (1) human involvement, (2) visibility, (3) a lack of a human safety net and (4) human conventions and limitations. These theories do not represent the full extent of the online experiences of the participants, but appeared important to their sense-making processes.

Expecting human involvement

The first folk theory, expecting human involvement, revolves around who, or what, one imagines interacting with online. A recurring example was that many participants expected discerning services provided by humans to be part of online shopping processes that are actually designed to reduce the role of people. Positive or negative experiences with shopping online were attributed to the personality traits of ‘people’ working in online stores, and interactions with the interface were described as ‘helpful’, ‘rude’, or ‘pushy’, rather than interpreted as automated responses or algorithmic recommendations.

Per (73) talked about ordering groceries for home delivery, describing the helpfulness of ‘she’ and ‘they’ online as rather unpredictable: Per:

I discovered that I had ordered eleven ketchup bottles! A second ‘1’ had crept in, but I thought they would realise that it was an error. Nobody needs eleven bottles. But what do you know, eleven bottles got delivered! Nobody stopped to think that this was totally off. (. . .) The same thing almost happened again, well. . . not with ketchup, of course, because my cupboard is full of that (laughs), but this time I realised it before I sent the order. I could not figure out how to correct it. . . I thought, ‘OK, I will just leave the order [not submit it] to allow her to get in touch with me and ask me about it’. But do you think she bothered getting in touch? And the next day, I had no delivery. That is too bad. She could have called me and asked.

Hilde (interviewer):

So, someone who works there monitors your online shopping?

Per:

Yes, for sure. She recommended things, too. . . and she was right. She suggested that I should buy eggs, and I needed eggs!

Per had encountered an algorithmic recommendation system for additional purchases, which he interpreted as an insightful person. Therefore, he expected ‘her’ to assist when he made mistakes and make follow-up telephone inquiries when he left the order hanging. This was his way of making contact, as he was unfamiliar with the chat function. Per understood algorithmic recommendations as helpful suggestions from a grocery shop employee. Interestingly, conversational agents and recommendation systems are often designed to communicate in seemingly natural language, which allows users to partly interact with information systems as they would with a person. Human-like interaction experiences can induce anthropomorphism (Seeger et al., 2021), which attributes human traits, emotions, and intentions to nonhuman agents (Epley et al., 2007). Therefore, Per’s expectations of human involvement are not surprising.

Bengt (88) tried to buy a book online and experienced a different interaction with the recommendation system: Bengt:

I only wanted this one book, but they kept showing me books and wanted me to buy them.

Hilde:

Who are ‘‘they’’?

Bengt:

I guess they work on commission because they were really pushing the sales. Those [people] working in the bookstore, they probably thought I was an old fool. I really did not appreciate that. They even put a book in the basket, my shopping basket.

Bengt described his experience as encountering a pushy rather than helpful agent. Per’s and Bengt’s accounts both attribute personality and intentions to algorithmic suggestions, which aligns with earlier research connecting the experience of intention to anthropomorphism (Nass et al., 1996). Individuals with limited technical knowledge of non-human agents are more likely to apply anthropocentric knowledge (Epley et al., 2007) with a subconscious intention to understand and logically explain events or circumstances involving non-human agents (Pfeuffer et al., 2019).

The theory of expecting human involvement draws on habitual expectations of encountering people, and attributing human traits to something human-like, building on experiences from the physical and social world when interpreting automated online interactions. The other three folk theories offer specific interpretations of such processes of transferring expectations from offline to online.

Expecting visibility

The second folk theory concerns expectations that one is visible to other humans when using online media and services. This theory encompasses a sense of place, envisioning the online world as an observable space, and a sense of the presence of others sharing the same space. Moreover, the theory demonstrates how the offline world functions as an interpretive device for understanding online spaces. It is a generalised theory of online space with practical and specific implications. This theory was widespread among our participants, although expressions of what visibility meant varied between different examples they talked about.

Anna (79) described how she planned to travel by finding ticket prices and departure times online, information she would formerly have obtained from a physical ticket office: Anna:

I often go back and forth on the webpage, looking for the correct information. And I am sure that the people working there are like, ‘What is she doing?’

Hilde:

They see you?

Anna:

Mhm . . . or not me . . . this [the curser] moves around, and I go up and down by using the arrow on my keyboard.

Anna compared navigating a webpage to moving around in a physical location, as though she were walking back and forth in the ticket office in a confused or suspicious manner. She presumed that others could observe her movements, implying they were paying attention to her. She described her curser as a representation of herself, followed by saying, ‘I try not to jump back and forth so much’, which suggests that her theorised visibility in the online space hindered her navigation.

Other participants voiced examples of how expectations of visibility influenced their actions. Ada (91) spoke of reading news online in terms of being observed. She reported reading ‘proper’ news first, despite being more interested in the royal family, to ensure that the editor would not take her ‘for a fool’. She was correct to assume that her choices as a digital news user will enter a feedback loop; however, she voiced a personal and human-centric interpretation of how this happened: ‘the editor’ would ‘see’ her and pass judgement on her choices, rather than her data being harvested and aggregated for analysis and personalised recommendations.

Similarly, Arvid (82) talked about visibility and privacy in online healthcare services: Arvid:

I might go on here [healthcare website or app] and find stuff, but I will call them [doctor’s office] if I want to book an appointment.

Hilde:

Why is that?

Arvid:

I think it is private. One does not have to include a lot of information, but others can still see that I have an appointment at that time, and I do not like that. In my time, we used to be more private, and I still am.

Arvid presumed that his online appointment would be visible to others, although it was unclear who this may be (e.g. other patients, medical staff, anyone online). Metaphorically, his caution could be understood as a reluctance to write his name on a noticeboard in the waiting room, or to speak to the health secretary with other patients waiting nearby. His approach demonstrated an awareness of privacy concerns mixed with uncertainty about what happens to his personal data, which is a well-known situation for internet users. In this case, knowledge of how a physical space worked helped him complete his task, because he could call the office instead and trust the person on the phone more than he would the platform.

The theory of expecting visibility envisions the online world as an observable space shared by others. Similar to the theory of human involvement, expecting visibility draws on metaphors. A metaphor is ‘understanding and experiencing one kind of thing in terms of another’ (Lakoff and Johnson, 1980: 5). Metaphors guide our comprehension of abstract concepts by allowing us to approach them in everyday language (Lakoff and Johnson, 1980). Metaphors such as cyberspace are conceptualised as a place where meaningful human activities occur (Markham, 2003). The internet cannot be perceived as a place only owing to its sense of architecture; it requires the presence of others, as the communal aspect of the context is the ‘focal point for this metaphorical framework’ (Markham, 2003: 8). Thus, there is a distinction between place and space. Our participants theorise the online world in terms of ‘place’, affecting their understanding of the online world as an observable place where they are visible and must act accordingly.

Expecting a lacking human safety net

The third folk theory concerns a feeling of being alone and vulnerable when using online services, thinking that someone will help if something goes wrong, and experiencing that this is not actually the case. While the above examples indicate expectations of a human safety net, the third folk theory highlights the partial realisation that this net cannot be trusted, and that users are alone. Several participants spoke of such gradual realisations. Here is one example:

What I do not like is that now we have to do all the jobs ourselves. I joke about being both a clerk and bank employee when I am on this [his Mac]. It used to not be like that; someone used to have these jobs. But now we are supposed to know everything ourselves? And we take on a great responsibility to make sure things are done correctly. Have you thought about that? I think about it a lot. Once, I booked a trip the wrong way around. That would never have happened if I had used a travel agent (Jan, 80).

While Jan pointed out failures in the safety net, connected to a critique of societal development and user responsibilisation, others had not come to terms with such expectations. Several struggled to understand their roles when using online services. Hans (82) recently moved houses and found the digital experience stressful. He had a startling realisation of online banking:

While I was moving, I did not have internet, and my computer was in a box somewhere. (. . .) It took like a month, maybe five weeks. . . and then I started to get letters. All my invoices had gone to debt collection. (. . .) Nobody notifies you about this, you see. I asked my bank advisor if she was sleeping on her job, but apparently, all transactions are automatic or digital now. That made me think. I used to think that if I did something wrong and entered the wrong number [when paying bills online], someone would notice and call me up. . . or if I forgot. But it turns out [that] I am all alone in that online bank (Hans, 82).

Hans had been using online banking for 2 years, believing that a safety net would prevent him from making mistakes. Although safety settings were available, they seemed to be unknown to him. Instead, he equated safety measures with having a human ‘on the job’ and naturally assumed that this responsibility would fall to a financial advisor in his bank. Realising the extent of automation made him worry about his competence and responsibilities.

Theresa (98) also struggled with online banking, and her daughter (in her 70s) had to take over. This was something they were both unhappy about: Theresa:

I used to go on the internet bank before.

Hilde:

Did you stop using it?

Theresa:

Yes, it turned out that I was making many mistakes. I did what my daughter told me to do, but then she said I did it wrong. I thought the people at the bank would make sure I did not do the wrong thing. So, I did as much as I could and left the rest for them. But I guess I was wrong in that, too (sighs).

Hans and Theresa trusted ‘someone’ who was not actually there, taking them from feeling somewhat comfortable experimenting with online banking to realising they were failing and also at risk. They could not understand how to reach customer assistance and tried to ask for help by leaving unfinished transactions, similar to Per’s example. A fundamental challenge is that when the different logic of online and offline services do not match, older adults’ folk theories could lead to passivity in situations that require user actions.

Expecting human limitations and social conventions

The fourth folk theory concerns attributing human limitations and social conventions to online services. This theory highlights what older adults expect online services to be capable of and which norms they attune their activities to, in both regards drawing on expectations transferred from offline situations.

Elsa (69) explained that she used Google on her smartphone. When asked if she also used Google on her Mac, she replied, ‘No, are they here as well? (. . .) I can’t believe they are here also. They must be very busy answering all our questions’. Later, the following exchange took place: Elsa:

There are often new words I do not know. I wrote down a list because I knew you were coming. Let’s see. . . podcast, streaming service, QR code. I wish I had a dictionary for the digital language.

Hilde:

Did you ever consider using Google?

Elsa:

[Appears surprised] Do you think Google will know?

Hilde:

You can Google most things. Even the weather. [Elsa had mentioned issues with a weather app earlier in the interview]

Elsa:

I thought Google was in America [the US]. How will they know anything about the weather here? [types in a search in English]

Hilde:

Have you ever googled in Norwegian? . . .. used Norwegian in your search?

Elsa:

In the question I ask Google? Never! I would consider that a bit rude.

Elsa’s perception of Google appears similar to a call centre with employees ready to answer questions, but only within a realm of expected human capability and knowledge. She theorised that she was communicating with English-speakers located in the United States. Although Google is an American tech company, her attribution of human limitations and social conventions restricted her effective use of search engines. Instead of searching for anything she needed, she carefully prioritised queries and formulated them politely in English.

While not all participants used search engines, those who did tend to formulate complete questions, including unnecessary words, to be polite. They wrote in English, even with limited language skills, while searching for Norwegian information. We thereby found prevalent logics of transferring human limitations from the offline world to online search. A few more explicitly compared search engines with humans answering individual questions: Hilde:

So, is it [online search] like an encyclopaedia?

Liv:

I would go to the library for that [encyclopaedia use], and I would ask for help. And I would ask the librarian if she could please look up ‘so and so’. I would not just say the word, like throw it out there. . . that would make no sense. It is the same here, I guess [points to computer screen]. I ask a question because that is how others know you are asking for something. . . looking for answers. It is quite simple! (Liv, 86)

Social conventions shape communication between people, and Liv transferred these conventions to online searches. Like Elsa, she presumed that Google is an English-speaking search engine, leading her to choose a Norwegian website (startsiden.no) instead. ‘Startsiden’ can be directly translated as ‘the start page’ and was popular in Norway in the 1990s as the starting point for navigating online. It operates as a highly commercial news aggregate and search site and was the preferred browser homepage for most participants. Consequently, the participants received sponsored search results. For instance, when searching for tax information, the first five results were ads for financial services, and the sixth result was for the official Norwegian tax office.

This folk theory highlights that applying offline habits as an interpretive device restricted the participants. They experienced language barriers, poorer search results, limitations in the type of information accessed, and self-imposed limitations on the number of searches. They made choices based on misleading presumptions. Socially accepted manners, politeness, and presumed intentionality were theorised by the participants as part of their interaction with search engines, which aligns with the results of earlier experimental research connecting these behaviours to anthropomorphism (Nass et al., 1996).

Conclusions and implications

With a framework of folk theories, we analysed older adults’ conceptions of the online environment, identifying four interconnected theories: expecting human involvement, expecting visibility, lack of a human safety net, and human imitations and social conventions. Through such folk theories, older adults draw on their lived experiences with the offline world as interpretative devices for interacting online, sometimes leading to misconceptions about how and when humans are involved. Instead of recommendation systems, they encounter ‘pushy salespeople’; instead of data protection, they think of personal observability; instead of safety features, they expect a person ‘on the job’; and instead of searching for information, they refrain out of politeness. Therefore, we find that the shared theme of transferring expectations of human presence sheds light on how older adults understand and use the internet. While DeVito et al. (2017) proposed a distinction between abstract and operational folk theories, the theories we identified are both: They are operational interpretations of how affordances in online services function, drawing on abstract assumptions of what it means to interact online, including nostalgic responses that mimic human interaction in offline spaces.

We will discuss some implications of our findings to the ‘digital literacy paradox’ (Schreurs et al, 2017) among older adults. Although their digital skills can be described as limited, our participants voiced values and concerns that are recognisable across population groups, as reported in similar studies on interpretations of online environments (Lomborg and Kapsch, 2020; Ytre-Arne and Moe, 2021). These concerns include reluctance to be observed, irritation at being misunderstood, and desire to protect privacy. However, the folk theories we have identified reveal considerable barriers towards effectively realising such intentions, not merely because of power structures in the digital environment, but also due to misconceptions that hinder the further development of digital literacies.

The older adults in our study have traversed the considerable barrier of actually going online, but are inexperienced users who need to develop their digital literacies further to use the internet in accordance with their interests. If we consider what they are likely to learn from their online experiences over time, the folk theories we have identified give a more detailed understanding of obstacles. First, the folk theory of expecting human involvement hinders critical awareness of how algorithms and recommendation systems work, obscuring knowledge of where on the internet one might expect to encounter such a system. Instead, the theory leads older adults to focus on presumed personality traits, which also limits the potential to transfer skills from one situation to another, as they would not expect the same person to work in different online stores. Second, the folk theory of expecting visibility depends on logics transferred from physical spaces, such as waiting rooms and ticket offices, in ways that are not conducive to developing a general understanding of the internet. Instead of freely exploring a website interface in their own time, they are reluctant to roam around too much, and hesitant to engage in experiences that advance learning. Third, the expectation of a human safety net affords passivity in situations where the user should be active, inducing withdrawal and feelings of vulnerability instead of succeeding in finding support. Fourth, the theory of expecting human social conventions is a barrier to effective searches for information, which also hinders older adults in using the internet to learn about the internet. We can therefore conclude that in order to advance digital literacies, older adults need stronger frameworks for transferring knowledge from one online situation to another rather than from offline to online.

Our findings also indicate how online services might be improved to work better for older adults: Design choices could consider that a task left unfinished might be a request for help, or that opportunities to search in different languages are not taken as given. Particularly, our analysis shows that older adults might benefit from online interfaces that are user-friendly but not confusingly human-like. While the participants expressed frequent worries about online fraud, they did not theorise human presence as a danger; rather, they drew on a sense of habitual nostalgia, where friendly interactions with fellow humans would invoke trust. Anthropomorphism increases trust in interfaces, as trust is based on cognitive and affective processes and interpersonal trait evaluations (Kulms and Kopp, 2019). Therefore, incorporating human-like features is a design approach to increase trust in non-human agents and accelerate the ease with which trust is established (Khan and Sutcliffe, 2014). Problems with these strategies are made visible as inexperienced users misconstrue the extent of human involvement.

We have contributed with a folk theory approach that brings to light underlying perceptions in how older adults make sense of the digital environment. Further research on different groups of older media users might benefit from further developing such approaches, also to understand more positive experiences with the internet. Regarding digital literacies, we underline the need to connect skill development to what is defined as important by the group in question, including their values and interests. This is in line with arguments from Nikou et al. (2020) that research should consider if and how digital technologies allow older adults to live their lives in ways that are valuable to them, and with Das’ (2023) emphasis on users’ abilities to champion one’s best interests as part of digital literacies. The folk theories we have identified, particularly the notions of transferring expectations and expecting human involvement, could be bolstered through more research towards a more extensive theory of how older adults understand the internet—underlining the value they place on human interaction.